WorksheetsMachine Learning-2
Total questions: 20
Worksheet time: 10mins
Which metric is most appropriate for evaluating a binary classification model?
Mean Squared Error (MSE)
Accuracy, Precision, Recall, and F1-Score
Root Mean Squared Error (RMSE)
R-Squared Error
Mean Absolute Error (MAE)
What does Precision measure in a classification model?
The proportion of actual positives that are correctly identified
The proportion of predicted positives that are actually correct
The proportion of actual negatives that are correctly identified
The overall accuracy of the model
How is the F1 Score calculated?
(Precision + Recall) / 2
2*(Precision *Recall) / (Precision + Recall)
Precision / Recall
(Precision - Recall) / (Precision + Recall)
(Precision + Recall) / (Precision -Recall)
How is Recall calculated?
TP / (TP + FN)
TN / (TN + FP)
TP / (TP + FP)
FN / (TP + FN)
(TP+TN)/(TP+TN+FP+FN)
If Precision is 0.8 and Recall is 0.6, what is the F1 Score?
0.74
0.72
0.70
0.68
0.66
Which metric is more important when false positives are costly?
Recall
Accuracy
Precision
F1 Score
False Positive Rate
Which of the following is true about Precision-Recall trade-off?
Increasing Precision always increases Recall
Increasing Recall always increases Precision
Increasing Recall may decrease Precision
Precision and Recall are independent of each other
If Precision is 1.0, what does it mean?
The model predicted all positives and negatives correctly
The model has a perfect Recall also
The model predicted all negatives correctly
The model predicted all positives correctly without any false positives
If a model has high Precision but low Recall, what does it indicate?
The model is predicting many false positives
The model is predicting many false negatives
The model is predicting many True negatives
The model is balanced
The model is overfitting
A classifier predicts 100 positive cases, out of which 80 are correct. The total actual positive cases are 90. What is the Recall?
80.8%
88.8%
92.2%
76.7%
80%
In a highly imbalanced dataset where positives are rare, which metric should be given more importance?
Accuracy
Precision
Recall
F1 Score
If a model has Precision = 0.5 and Recall = 1.0, what is the F1 Score?
0.50
0.67
0.75
0.33
1.00
If a model has Precision = 1.0 and Recall = 0.5, what does it mean?
The model predicted all positive cases correctly but missed some actual positives
The model predicted all positives correctly and detected all actual positives
The model is overfitting
The model has a high F1 Score
The model is underfitting
Why is F1 Score preferred over Accuracy for imbalanced datasets?
F1 Score is always higher than Accuracy
Accuracy does not consider false positives and false negatives equally
F1 Score is easier to compute
Accuracy is the best metric for imbalanced datasets
What happens if K in KNN is too small?
The accuracy increases significantly
The model ignores outliers
The model becomes too simple and may underfit
The model becomes too complex and may overfit
Which distance metric is most commonly used in KNN?
Manhattan Distance
Cosine Similarity
Euclidean Distance
Jaccard Distance
Hamming Distance
Which of the following is NOT true about KNN?
It is a non-parametric algorithm
It can be used for both classification and regression
It requires training before making predictions
It is sensitive to the choice of distance metric
Naïve Bayes is most suitable for which type of dataset?
Large dataset with correlated features
Small dataset with independent features
Large dataset with independent features
Small dataset with correlated features
What is the role of the prior probability in Naïve Bayes?
It represents the probability of a class before considering the features
It is calculated after observing the data
It is always 0.5 for binary classification
It does not affect the classification
What kind of probability does Naïve Bayes compute?
Joint probability
Conditional probability
Prior probability
Posterior probability
